napistu 0.1.0__py3-none-any.whl → 0.2.4.dev2__py3-none-any.whl

napistu/modify/pathwayannot.py

@@ -24,6 +24,8 @@ from napistu.constants import ONTOLOGIES
 from napistu.constants import ENSEMBL_PREFIX_TO_ONTOLOGY
 from napistu.modify.constants import COFACTOR_SCHEMA
 from napistu.modify.constants import COFACTOR_CHEBI_IDS
+from napistu.modify.constants import NEO4_MEMBERS_SET
+from napistu.modify.constants import REACTOME_CROSSREF_SET
 
 logger = logging.getLogger(__name__)
 
@@ -499,12 +501,14 @@ def add_entity_sets(
     reactome_members = _read_neo4j_members(neo4j_members)
 
     # create missing species and compartmentalized species
+    logger.info("Adding entity set species")
     (
         merged_membership,
         new_species_for_sbml_dfs,
         set_component_species_ids,
     ) = _add_entity_sets_species(sbml_dfs, reactome_members)
 
+    logger.info("Adding complex formation species")
     (
         new_compartmentalized_species_for_sbml_dfs,
         updated_compartmentalized_membership,
@@ -515,6 +519,7 @@ def add_entity_sets(
         set_component_species_ids,
     )
 
+    logger.info("Adding entity set reactions")
     (
         new_reactions_for_sbml_dfs,
         new_reaction_species_for_sbml_dfs,
@@ -569,7 +574,8 @@ def add_reactome_identifiers(
 
     """
 
-    select_reactome_ids = _format_reactome_crossref_ids(crossref_path)
+    logger.info("Reading Reactome crossref ids")
+    select_reactome_ids = _read_reactome_crossref_ids(crossref_path)
 
     # read all current identifiers
     current_ids = sbml_dfs.get_identifiers(SBML_DFS.SPECIES)
@@ -582,6 +588,7 @@ def add_reactome_identifiers(
     )
 
     # combine existing s_ids with additional cross-ref annotations using uniprot ids
+    logger.info("Merging Reactome crossref ids with existing identifiers")
     merged_crossrefs = _merge_reactome_crossref_ids(
         current_molecular_ids, select_reactome_ids
     )
@@ -637,7 +644,8 @@ def add_reactome_identifiers(
     updated_identifiers.index.name = SBML_DFS.S_ID
     updated_identifiers.name = "new_Identifiers"
 
-    # add new identifiers to species table
+    # add new identifiers to species tabl
+    logger.info("Adding new identifiers to species table")
     updated_species = sbml_dfs.species.merge(
         updated_identifiers,
         left_index=True,
@@ -1173,6 +1181,9 @@ def _read_neo4j_members(neo4j_members: str) -> pd.DataFrame:
         with bfs.open(path, "rb") as f:
             reactome_members = pd.read_csv(f).assign(url="")
 
+    # check that the expected columns are present
+    utils.match_pd_vars(reactome_members, NEO4_MEMBERS_SET).assert_present()
+
     reactome_members[IDENTIFIERS.ONTOLOGY] = reactome_members[
         IDENTIFIERS.ONTOLOGY
     ].str.lower()
@@ -1295,9 +1306,13 @@ def _merge_reactome_crossref_ids(
     ].drop("_merge", axis=1)
 
     merged_crossrefs = pd.concat([uni_rct_with_crossrefs, uni_no_rct_with_crossrefs])
-    assert (
-        uni_rct_with_crossrefs.shape[0] + uni_no_rct_with_crossrefs.shape[0]
-    ) == merged_crossrefs.shape[0]
+    if (
+        not (uni_rct_with_crossrefs.shape[0] + uni_no_rct_with_crossrefs.shape[0])
+        == merged_crossrefs.shape[0]
+    ):
+        raise ValueError(
+            "The number of merged crossrefs does not match the sum of the number of uniprot + reactome crossrefs and uniprot-only crossrefs"
+        )
 
     species_with_protein_and_no_gene = current_molecular_ids[
         current_molecular_ids[IDENTIFIERS.ONTOLOGY] == ONTOLOGIES.UNIPROT
@@ -1332,9 +1347,9 @@ def _merge_reactome_crossref_ids(
     return merged_crossrefs
 
 
-def _format_reactome_crossref_ids(
+def _read_reactome_crossref_ids(
     crossref_path: str,
-) -> str:
+) -> pd.DataFrame:
     """
     Format Reactome CrossRef IDs
 
@@ -1347,7 +1362,7 @@ def _format_reactome_crossref_ids(
 
     Returns
     -------
-    select_reactome_ids: str
+    select_reactome_ids: pd.DataFrame
         Crossref identifiers
 
     """
@@ -1357,6 +1372,9 @@ def _format_reactome_crossref_ids(
         with bfs.open(path, "rb") as f:
             reactome_ids = pd.read_csv(f)
 
+    # check that the expected columns are present
+    utils.match_pd_vars(reactome_ids, REACTOME_CROSSREF_SET).assert_present()
+
     # only use ensembl and pharos for now
 
     # rename pharos ontology

napistu/network/neighborhoods.py

@@ -84,7 +84,8 @@ def find_and_prune_neighborhoods(
 
     if isinstance(compartmentalized_species, str):
         compartmentalized_species = [compartmentalized_species]
-    assert isinstance(compartmentalized_species, list)
+    if not isinstance(compartmentalized_species, list):
+        raise TypeError("compartmentalized_species must be a list")
 
     if isinstance(precomputed_distances, pd.DataFrame):
         logger.info("Pre-computed neighbors based on precomputed_distances")
@@ -832,8 +833,8 @@ def add_vertices_uri_urls(
 
     """
 
-    assert isinstance(vertices, pd.DataFrame)
-    assert vertices.shape[0] > 0
+    if vertices.shape[0] <= 0:
+        raise ValueError("vertices must have at least one row")
 
     # add uri urls for each node
 
@@ -880,7 +881,8 @@ def add_vertices_uri_urls(
             [neighborhood_species_aug, neighborhood_reactions]
         ).fillna("")
 
-    assert isinstance(updated_vertices, pd.DataFrame)
+    if not isinstance(updated_vertices, pd.DataFrame):
+        raise TypeError("updated_vertices must be a pandas DataFrame")
     if vertices.shape[0] != updated_vertices.shape[0]:
         raise ValueError("output vertices rows did not match input")
 
@@ -1532,7 +1534,8 @@ def _calculate_path_attrs(
         raise TypeError("neighborhood_paths should be a list of lists of edge indices")
     if not isinstance(vertices, list):
         raise TypeError("vertices should be a list of list of vertices")
-    assert len(vertices) > 0  # control for length zero vertices upstream
+    if len(vertices) <= 0:
+        raise ValueError("vertices must have length greater than zero")
     if len(neighborhood_paths) != len(vertices):
         raise ValueError("vertices and neighborhood_paths were not the same length")
 
@@ -1588,7 +1591,13 @@ def _calculate_path_attrs(
     path_attributes_df = pd.concat([path_attributes_df, edgeles_nodes_df])
     neighborhood_path_entities.update({x: {x} for x in edgeless_nodes})
 
-    assert path_attributes_df.shape[0] == len(neighborhood_paths)
-    assert len(neighborhood_path_entities) == len(neighborhood_paths)
+    if path_attributes_df.shape[0] != len(neighborhood_paths):
+        raise ValueError(
+            "path_attributes_df row count must match number of neighborhood_paths"
+        )
+    if len(neighborhood_path_entities) != len(neighborhood_paths):
+        raise ValueError(
+            "neighborhood_path_entities length must match number of neighborhood_paths"
+        )
 
     return path_attributes_df, neighborhood_path_entities